Mari’s research on musicians and dancers

In my studies, I record musicians and dancers simultaneously. This article will present a rhythm study of a specific style of traditional Norwegian dance music called telespringar. It is based on a motion capture recording of three telespringar performers: one fiddler and two dancers.

Musical rhythm

The experience of musical rhythm often involves an interaction between sounding rhythms and endogenous reference structures, such as pulse and meter. This is typically the level of the musical rhythm that you will tap your foot or nod your head to. Whereas the pulse refers to successive beats, the meter could be described as a grouping of those beats. For example, in “Happy Birthday to You,” the pulse beats are grouped into three. In musical notation, the meter is indicated by the time signature, such as 4/4 or 3/4, at the beginning of the piece of music.

Meter and motion

I am particularly interested in the relationship between the musical meter and performers’ periodic body motion. Some years ago, I investigated the meter in telespringar by analyzing performers’ periodic body motion in a telespringar performance: the fiddler’s foot stamping and the dancers’ vertical body motion.

Set-up

The recordings were carried out in the fourMs Lab using a nine-camera motion capture system from Qualisys. The system tracked the motion of reflective markers at a frame rate of 200 Hz. I recorded video files together with the motion capture data for reference. To ensure high-quality audio, I also recorded audio separately in Logic Pro X.

To be able to synchronize the motion and sound recordings, I used a clipboard with markers attached. Such a “movie-style” clipboard functioned well. It was easy to detect the simultaneous spikes in the audio recording and the acceleration plots of the motion capture data. Subsequently, these spikes were used as a common reference point when aligning the data streams.

Marker placement

The markers were primarily attached to the participants’ joints. When I record several participants simultaneously, I find it useful to introduce some supplementary control markers. These markers are not intended for analysis but for distinguishing the participants. In folk dancing, for example, the dancers twirl around and hold one another’s arms in different ways. Without control markers attached to the dancers’ torsos (front and back) and upper and lower arms, it can be tough to tell them apart. The control markers can also be connected with the proximate joint markers by “bones, “ facilitating the identification process.

In this particular study, I recorded one fiddler and two dancers. In addition to markers placed on the participants’ bodies, I decided to attach four markers to the fiddle and two to the bow. Below is a figure showing the placement of the markers attached to the dancers, fiddler, fiddle, and bow. The markers have been identified and connected by “bones.”

You can watch a short excerpt of the motion capture recording in this video:

Ecological validity

A motion capture lab is not a typical environment for folk music and dancing. The attachment of reflective markers to one’s body can exacerbate the feeling of being in an artificial environment even more. I am interested in the participants performing as they would normally do. Hence, the performers must be comfortable. Because of this, I chose to use only the motion capture jacket and the motion capture hat. For the legs and feet, I used straps and tape to attach the markers. I also dimmed the light slightly.

Analysis and results

According to existing hypotheses and findings, the fiddler’s foot stamping and the vertical motion of the dancers’ center of gravity, the libration pattern, might be related to the meter in telespringar. Here, the fiddler’s foot stamping revealed unambiguous acceleration peaks related to the points in time when the feet (toes) hit the floor. The duration pattern based on the fiddler’s foot stamping revealed a stable long–medium–short pattern, indicating that the metrical structure in telespringar is asymmetrical.

The dancers’ libration curves were based on the vertical position of the markers placed on their right hip. I divided the dancers’ libration curves according to the fiddler’s foot stamping. This showed that the shape of the curves was very stable and that the beginning of a measure seemed to fall between turning points. The duration of the first beat seemed to correspond to a small vertical “S-shape”; the duration of the second beat to a deep down/up motion; and the duration of the third beat to an up/down motion.

Below is a figure showing plots of (a) the dancers’ vertical hip motion and (b) the vertical acceleration of the fiddler’s foot stamping. All the motion trajectories are chunked into segments of two measures and plotted on the same graph. In the figure, you can see how the fiddler’s foot stamping follows a regular long–medium–short duration pattern. The mean beat positions, estimated using the fiddler’s foot-stamping, are indicated by vertical dashed lines in the plots.

Summary

My study concluded that the meter in telespringar is inherently asymmetrical. Both the fiddler and the dancers seem to relate to a shared reference, indicated by the fiddler’s foot stamping. This shared reference also corresponds to the shape of the dancers’ libration curves.

References

• FourMs Lab (2016). Motion Capture – Norwegian Telespringar. YouTube video, June 10. https://youtu.be/oMO7BhzCHa0
• Haugen, M. R. (2016). Investigating Periodic Body Motions as a Tacit Reference Structure in Norwegian Telespringar Performance. Empirical Musicology Review, 11(3–4), 272—294. doi:10.18061/emr.v11i3-4.5029
• Haugen, M. R. (2017). Investigating Musical Meter as Shape: Two Case Studies of Brazilian Samba and Norwegian Telespringar. In E. Van Dyck (Ed.), Proceedings of the 25th Anniversary Conference of the European Society for the Cognitive Sciences of Music (pp. 67—74). Ghent, Belgium: University of Ghent.